ISO Standards for Tea
Intergovernmental Group- on Tea 10th Session New Dehli, 12-14 May 2010
Andrew Scott, Chairman ISO TC/34 SC 8 Tea 1 2 Tea
Over 3 million tonnes produced
In over 30 countries
Providing over 1000 billion cups of tea
3 International Standards for Tea
International standards are vital to facilitate international trade
To ensure consumer expectations are met
To provide guidance and common understanding of Good Manufacturing practices via compositional specifications
To provide validated methods of analysis
4 International Standards for Tea 1970’s - Tea Committee became ISO Working Group 8 - ISO Technical Committee 34 Sub-committee 8 – Tea
1977 - ISO 3720 Black tea standard 8 testing methods to measure basic tea parameters
1980 - ISO 1839 Sampling tea
1980 - ISO 3103 Preparation of liquor for sensory analysis
1982 - Glossary of terms
1990 - Instant tea standard 4 supporting test methods
5 …. why measure polyphenols in tea?
6 ISO BLACK TEA SPECIFICATION - ISO 3720 1977 (First edition)
“INTRODUCTION
The content of CAFFEINE and the content of POLYPHENOLIC constituents are important chemical characteristics of black tea, but it has not been possible to include limits for either of them in the specification. In the case of CAFFEINE, agreement has not yet been reached on a standard method for the determination.
In the case of POLYPHENOLIC CONSTITUENTS, knowledge about test methods is not sufficiently developed to justify the standardization of any one of the methods in existence; moreover, information on the content of these
7 constituents is only available for a few types of tea.” Tea & International Standards
1995 - Measurement of caffeine in tea and instant tea
- Measurement of caffeine in tea, instant tea and decaffeinated tea
2005 - Measurement of total polyphenols in tea
2005 - Measurement of catechins in green (and black) tea
8 Substances Characteristic of Tea (15th Meeting ISO/TC 34/SC 8, 1992)
Objectives
Further development of methods for catechins and total polyphenols in green teas
Obtain catechin and total polyphenols data on a range of green teas representative of those in international trade
Examine the applicability of these methods for black and oolong teas to facilitate discrimination from green
Also
Catechin data is needed for the study of the health benefits of tea 9 Measuring Polyphenols
The methods should be:
- Capable to be used by all scientists in producer & consumer countries
- Produce reproducible results within a laboratory
- Produce reproducible results between laboratories
- A validated method of analysis
10 Measuring Total Polyphenols Folin Denis reagent established as phenol reagent to quantify tyrosine in proteins, vanillin in vanilla and method of choice for estimating mono- and poly- phenolic molecules in plant foods.
Issues:
- phenols approximately 50% ionised at pH 9-10 - precipitate with increasing reagent or sodium carbonate concentration or both - Beer Lambert Law not strictly followed - saturated sodium carbonate: variation with temperature, equilibration not rapid - Timing of adding reagents important - Temperature and other factors affect the blue colour formation - distilling at high altitude (Kenya tea estate) not equivalent to London / Colworth - empirical method
First trial evaluated the Folin Denis reagent:
– variability due to making up the reagent 11 Measuring Total Polyphenols Folin Ciocalteu reagent similar to Folin Denis reagent and is available commercially
Advantages FC over FD:
- no precipitation except in excess potassium
- more intense and more stable blue colour
- no difficulties in preparation – off the shelf.
- gives less colour in proportions to that of FD for possible interfering reductants [high content of active oxidant in FC promoting more complete oxidation and measurement of more slowly reacting phenolics?]
12 Measuring Total Polyphenols
Agreed to base test on a method published by Singleton & Rossi (1965) using the Folin-Ciocalteu reagent
Test was developed by a working group of 4 laboratories
Trials:
#1 2 black (Kenya & Argentina) and 1 green (China) teas
#2 2 black (Kenya & Argentina) and 1 green (China) teas
13 Measuring Total Polyphenols First International ring trial - 20 participating laboratories:
- showed promising results - identified that training was required for good precision - improvements to protocol required - add a training set of samples
Issues:
- sodium carbonate solution: - saturated (Folin Denis) to 20% (Singleton & Rossi 1965) - high blanks - mixing of reagents - timings of additions - colour development time (FC vs FD) - age & batch of reagent differences? - laboratory temperature 14 - extraction - volume variation / extraction recovery for catechins & flavonoids? Measuring Total Polyphenols
Extraction:
Total Polyphenol & Catechin Analysis
- tea must be finely ground (< 0.5mm)
- small but representative sample of ground tea (0.2g)
- exhaustive double extraction with hot (70C) 70% methanol give high yield and low catechin degradation
- extract samples are stable at 4C for 24 after preparation
Note:
- disposable plastic tubes are easier and more reliable to use than re-useable glassware to eliminate trace contamination etc. 15 Measuring Total Polyphenols
Second International ring trial - 24 participating laboratories:
- within laboratory repeatability without exception improved
- between laboratory reproducibility values similar to 1st trial
16 Measuring Total Polyphenols
y = 0,0132x + 0,0113
17
Green leaf tea - China To 30 tal Mean 24.5 Po 25 ly ph en 20 ol s 15 % on D 10 M 5
0 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 Laboratory Number
Black leaf tea - Kenya Tot al 30 Pol Mean 18.8 yp 25 he nol s 20 % on 15 DM
10
5
0 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 Laboratory Number
Tot Black leaf tea - Argentina al Pol 30 yp Mean 13.9 he 25 nol s 20 % on DM 15
10
5
0 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 Laboratory Number 18 Measuring Total Polyphenols
19 Measuring Total Polyphenols Total extraction of polyphenols:
ISO Black tea – 14.63% on dry matter basis (54 samples) Green tea – 17.29% on dry matter basis (102 samples)
University of Braunschweig study
Black teas: UK 14.92% Intl A 14.61% Intl B 15.14% USA 15.29% UK Tea Trade database:
557 mg/l polyphenols in a brewed cup of tea (71 samples)
20 21 Measuring Catechins
The method should be:
- Capable to be used by all scientists in producer & consumer countries
- Fit for the purpose
- Produce reproducible results within a laboratory
- Produce reproducible results between laboratories
22 Measuring Catechins Uses HPLC to separate the constituents and measures the UV absorbance of the individually separated catechins
Samples – Stage 1 instant teas – 1 green & 1 black - Stage 2: 2 green teas [China sm lf & China bold lf] 2 black teas [Darjeeling Snowview & Lingia]
Remove all variation other than laboratory testing:
- the same columns used (Unilever prep) [Phenomenex Luna® 5m Phenyl-Hexyl 250x4.6mm with Phenomenex Security Guard® 4x3mm Phenyl-Hexyl guard]
- pure catechin standards used (Unilever)
23 - tea samples ground to powder and mixed to homogeneity Measuring Catechins Extraction - same as Total Polyphenol Analysis
HPLC conditions
- HPLC with Luna® Phenyl-Hexyl column improves resolution of catechins compared with C18
- UV wavelength detection (278 nm)
- Temperature control to 35C to maintain resolution and prevent drifting peaks
-Solvent gradient 2% acetic acid - acetonitrile with EDTA; Acetonitrile 9-32% over 15mins
- Samples for analysis are diluted in stabilising solution containing EDTA and ascorbate in acetonitrile
- Oxidation (metal catalysed or otherwise) must be avoided for reproducible results in dilute solutions 24 Measuring Catechins
Test was developed by a working group of 4 laboratories
#1 International ring trial - 17 laboratories participated :
16 results were returned Issues:
- adequate within laboratory - variability between laboratories
- recalculation of data using consensus relative response factors improved repeatability
- RRF’s viable approach for quantitation of catechins [pure catechins – not readily available, impure and very costly]
- variability most likely due to degradation during chromatography (metal 25 contamination) -> include EDTA in mobile phase Measuring Catechins
#2 International ring trial - 17 laboratories participated:
14 results were returned
Statistical methods: Cochrane test (duplicates), Dixon test (between laboratories), Friedman test (consistent highs/lows)
Relative response factor relative to caffeine std (more convenient retention time than gallic acid)
Standard purity assessments [ HPLC, NMR, moisture]
26 Measuring Catechins
Comparison of statistical analysis results on recalculated RRF’s vs Laboratory’s calibrations:
- total catechin values lower as a consequence of using standard RRF’s expressed on dry matter basis
- an improvement on repeatability values with RRF recalculated data set
- a marginal improvement in reproducibility in 3 out of the 4 teas with the RRF re-calculated data set.
27 28 29 30 Green Leaf Tea - Sample 1 Green Leaf Tea - Sample 2
25 25 Mean 14.0 Mean 17.4
20 20
15 15
10 10
%CatechinsDMon %CatechinsDMon 5 5
0 0 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 Laboratory No. Laboratory No.
Black Leaf Tea - Sample 3 Black Leaf Tea - Sample 4
25 25 Mean 10.8 Mean 8.2 20 20
15 15
10 10
%CatechinsDMon %CatechinsDMon 5 5
0 0 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 Laboratory No. Laboratory No.
31 Measuring Catechins
32 Measuring Catechins
33 Measuring Catechins
34 Measuring Catechins
Total catechin content:
ISO Black tea 1 – 8.93% on dry matter basis Black tea 2 – 6.81% on dry matter basis Green tea 1 – 12.14% on dry matter basis Green tea 2 – 14.78% on dry matter basis
Within laboratory repeatability of the analysis is good (9)
Between laboratory reproducibility has larger variation but this is to measure 5 constituents (9)
35 Measuring Catechins
Conclusions:
- method adequately repeatable within laboratory
- largest contribution to total variance comes from between laboratories [analysis of 7 constituents]
- recalculation of data using consensus RRF’s improved the repeatability but only a marginal improvement on reproducibility
- variance from some laboratories - degradation of catechins - take precautions [EDTA already added to mobile phase]
36 37 Conclusions
Two Internationally validated methods analysis for measuring (i) total polyphenols and (ii) catechins in leaf or instant tea are now available.
These methods can be adapted to measure polyphenol & catechins in tea brews
38 Recommendation
Commend these robust methods of analysis for routine use in the measurement of polyphenols in tea.
Tea trade & academics in each country goes about collecting data on teas available in the trade to generate typical values of polyphenols in tea
Agree with regulatory authorities the appropriate use of typical polyphenol content in teas with consumers (marketing, on pack etc)
39 References
Singleton, V.L., Orthofer,R. and Lamuela-Raventos, R.M. Analysis of total polyphenolics and other oxidation substrates and antioxidants by means of Folin-Ciocalteu reagent. In: Methods in Enzymology. Oxidants and Antioxidants, Part A,Lester Packer, ed. (1999), 299, pp. 152-178.
ISO/TC 34/SC 8 – Tea N513, Report from the project leader of the ISO ring test on ISO/CD 14502-2 (Tea – Methods for determination of substances characteristic of green and black tea – Part 2: Determination of catechins in green tea – Method using high performance liquid chromatography).
40 Current Projects
Publication of polyphenol data in green and black tea
Green tea specification
Black tea specification revision
White Tea specification
Theanine measurement in tea
41 Acknowledgements
Dr Peter Collier ex-Chairman ISO TC 34/SC 8 - Tea
Clive Dacombe - Unilever Professor Ulrich Engelhardt – University of Braunschweig Katie Donnelly – Tetley Dr Andrew Lea – RSSL
Dr Conrad Astill - Unilever Dr Jacek Obuchowicz
42 43